Fuel 214 (2018) 330–341

Contents lists available at ScienceDirect

Fuel
journal homepage: www.elsevier.com/locate/fuel

Full Length Article

Experimental investigation of the effects of ammonia solution (NH3OH) on T

engine performance and exhaust emissions of a small diesel engine
⁎
Zehra Şahina, , İmdat Ziya Akcancaa, Orhan Durgunb
a
Karadeniz Technical University, Faculty of Engineering, Mechanical Engineering Dep., Trabzon, Turkey
b
Avrasya University, Mechanical Engineering Dep., Trabzon, Turkey

A R T I C L E I N F O A B S T R A C T

Keywords: In the present study, the effects of ammonia fumigation on the engine performance and exhaust emissions have
Ammonia injection into intake air (ammonia been investigated experimentaly in a small diesel engine. Experiments have been performed for (2, 4, 6, 8, 10)%
fumigation) (by vol.) ammonia ratios at different engine speeds and loads. Here, 25% ammonia solution (25% am-
Exhaust emissions monia + 75% water) have been used and this solution has been injected into intake air by a carburetor, which
Engine characteristics
main nozzle section is adjustable. The test results showed that brake specific fuel consumption (bscf) increases at
Diesel engine
2000 and 3000 rpms, but it decreases at 2600 rpm. The maximum reduction of bsfc has been determined as
7.28% for 5.48% ammonia ratio at 2600 rpm. Effective efficiency increases at all of the selected engine speeds
and loads. However, the increase ratios of effective efficiency at 2600 rpm are higher than that of the other
engine speeds. Exhaust emissions have been measured at 2200 rpm for different ammonia ratios and different
loads and at 2600 rpm for different ammonia ratios, under only 6 Nm loads. It has been determined that carbon
dioxide generally reduces at selected two engine speeds. However, nitrogen oxides, hydrocarbons and carbon
monoxide generally increase. Total fuel cost for applied different ammonia ratios becomes lower than that of
neat diesel fuel at 2600 rpm. At the other selected engine speeds; although total fuel cost decreases for some
working conditions, it increases in some operating conditions. The maximum reduction of total fuel cost has been
determined as 8.87% for 5.48% ammonia ratio at 2600 rpm.

1. Introduction contributes to increase global temperature and it is regarded as

Instead of have being used alcohol and vegetables fuels in internal
combustion engines, petroleum based fossil fuels have been the main
fuels of these engines for nearly 130 years. Petroleum based fuels are
being used in different areas and they have a significant place in our
country’s economy as well as in main other countries. In recent years,
the risk of declining of petroleum sources, the increasing of environ-
ment pollution and also the rising of the levels of greenhouse gases
force both governments and industries worldwide to mitigate the con-
sumption and harmful effects of petroleum fuels. Especially the increase
of environment consciousness stimulates the interest of researchers and
engine manufacturers to search for alternative energy sources such as
LPG, biodiesel, DME, natural gas and alcohol fuels. Among these,
oxygenated fuels are critical importance due to their ability to reduce
exhaust emissions and also increase engine performance [1–3].
Besides, oxygenated fuels, especially biodiesel and ethanol, can
form a CO2 life cycle and have a neutral effect on total carbon dioxide
(CO2) [4,5]. Although CO2 is known a complete combustion product, it
greenhouse gases. It is estimate that CO2 will increase with approxi-
mately 40% from the year 2007 to 2030 [6]. Therefore, reduction of
CO2 is as important as the reduction of the other exhaust emissions. It is
known that ammonia and hydrogen, produced from renewable re-
sources, can also be considered as carbon-free fuels [7–10]. Hydrogen is
the most interesting fuel of the future, and it has been studied ex-
tensively in the literature [9–12]. However, the number of the studies
on ammonia are limited compared to hydrogen and also the other al-
ternative fuels [7]. In fact, ammonia has been used in busses in Belgium
in 1942 during World War II [6,10]. Later, in the 1960’s the U.S. Army
also performed both theoretical and experimental studies to test the
feasibility of ammonia as an alternative fuel source for internal com-
bustion engines [6,10]. After the aforementioned, it can be said that
ammonia studies as engine fuel are very limited until recent years.
Ryu et al. [8] investigated the engine performance of a spark igni-
tion engine using direct injection of gaseous ammonia. In this study,
gasoline was used as a second fuel to enhance combustion and an ap-
propriate ammonia injection strategy was developed to obtain best

⁎
Corresponding author.
E-mail addresses: zsahin@ktu.edu.tr (Z. Şahin), akcanca@gmail.co (İ. Ziya Akcanca), odurgun@ktu.edu.tr (O. Durgun).

https://doi.org/10.1016/j.fuel.2017.10.034
Received 20 March 2017; Received in revised form 6 June 2017; Accepted 5 October 2017
Available online 21 November 2017
0016-2361/ © 2017 Elsevier Ltd. All rights reserved.
Z. Şahin et al.
engine performance. The test results showed the engine power increases
by advancing ammonia injection timing. Also, the brake specific energy
consumption with ammonia was very similar to that with neat gasoline.
However, this study showed that the using ammonia increased nitrogen
oxide (NOx) and hdycarbon (HC) emissions and decreased carbon
monoxide (CO) emission. After this study, Ryu et al. [9] studied the
effect of combustion of hydrogen generated by an ammonia dissociation
catalyst on engine performance and exhaust emissions in the same
spark ignition engine. In their study, it was found that the conversion of
ammonia into hydrogen increased engine power significantly and re-
duced fuel consumption, CO, HC, NH3 and NOx emissions considerably.
This study showed that the ammonia dissociation catalyst can enable to
use ammonia as a hydrogen carrier in internal combustion engines ef-
fectively [9]. Actually, ammonia is easily liquefied and stored under
ambient temperature of 300 K at a pressure of approximately 10 bar and
it is relatively easy to store as liquid or gas. Also, its infrastructure is
well established and it can be regarded as a hydrogen carrier [7,10].
Although ammonia has high self-ignition temperature, the use of
ammonia in diesel engines have also examined by using different
techniques [7,10,12–14]. Anhydrous ammonia was investigated ex-
perimentally and theoretically in Cooperative Fuel Research (CFR)
engine by Starkman et al. [14]. In this study, a new injection system
and also a spark ignition system were added to their test engine. Results
indicated that ammonia decreased engine power and it increased bsfc
approximately (1–1/2) fold. This study showed that liquid ammonia
may be burn satisfactorily as a fuel in diesel engine at normal com-
pression ratios by using spark ignition to ignite the injected fuel. Later,
Reither et al. [10] were investigated the use of ammonia by applying
dual-fuel approach or fumigation method, by making a minor mod-
ifications in normal diesel engine. In their study, ammonia was injected
in the intake air (AF) by using an adapted injection system as an ad-
ditional fuel and various combinations of ammonia and diesel fuel were
successfully tested. From their experimental results showed that AF
increased HC, CO, NOx and bsfc. On the other hand, it reduced soot
emissions significantly. Another ammonia fumigation study was per-
formed by Gill et al. [12]. In their study, ammonia gas, dissociated
ammonia (a mixture of H2, N2 with small percentages of NH3) and pure
hydrogen was injected into intake air in a diesel engine and they found
that by substituting only 3% of the intake air by ammonia, the diesel
fuel consumption and the CO2 decreased at the levels of 15%. Also, they
determined that AF increased HC, NOx emissions but it decreased CO
emission. There are also different fuels – ammonia mixtures studies in
literature. Such a study was performed by Ryu et al. [10]. They was
tested various mixtures of ammonia (NH3) and dimethyl ether (DME) in
a diesel engine. The original diesel fuel injection system of their test
engine was replaced with a new system for injecting ammonia-DME
mixtures into the cylinder directly. Emissions data showed that soot
emissions remained extremely low for the fuel mixtures tested. On the
other hand, as ammonia concentration increased, CO and HC and NOx
emissions increased. It can be understood from the above informations
that the use of ammonia generally increases fuel consumption and CO,
HC and NOx emissions and it reduces CO2 and soot emission.
For the fulfilling of international emission standards of internal
combustion engines, after-treatment technologies are also used apart
from alternative fuels and combustion improvement studies etc.
[15,16]. For example, various additives, patented by various compa-
nies, are used in exhaust systems of diesel engines which used on truck,
buses and passanger cars. These additives are mixtures of water and
urea which have similar structure and specific characteristics. This
mixture is decomposed into exhaust gases and transformed into am-
monia. This ammonia reacts with the incomplete combustion products
such as NOx, CO and HC coming out of the engine and it converts them
to complete combustion products such as N2 and H2O and CO2. It was
thought by authors that ammonia can improve combustion process and
thus it can prevent to produce incomplete combustion products by its
using as an additive considering its the above-mentioned feature.
331
Fuel 214 (2018) 330–341
Evaluating of these effects of ammonia, it is intended to investigate the
use of ammonia in a small diesel engine. Furthermore, it can be seen in
the relevant literatures that detailed experimental studies on ammonia
include cost analysis are not reported. Furthermore, in recent years,
studies on ammonia have also been increased. As a result, ammonia as a
fuel is becoming an interesting area for researchers.
It is well known that ammonia is in the gas phase under normal
ambient air conditions. Consequently, there may be arisen some diffi-
culties for using gas ammonia in engines. For this reason, ammonia
solution (25% ammonia + 75% water) was used in the present study.
Ammonia solution does not mix with diesel fuel homogeneously.
Furthermore, due to the poor auto-ignition capability of ammonia and
corrosive effect of water content on fuel system, the use of diesel-am-
monia solution blends are not feasible [17,18]. Hence, ammonia solu-
tion was injected into intake air by using an adapted simple carburetor
and diesel fuel injected on this mixture at end of the compression
process. Here, at the application of this method any change has not
been made in original injection system of the test engine and the engine
mainly operates due to diesel principle. As a result, as above explained,
the main objective of the present work is to investigate experimentally
the effects of addition of ammonia solution at low ratios such as (2, 4, 6,
8, 10%, by vol.) on performance and emission characteristics of a small
diesel engine at different loads and engine speeds. Furthermore, cost
analysis has been conducted and the obtained ammonia solution results
were compared to NDF.
2. Experimental study
2.1. Experimental apparatus
The test engine used in the present study is one-cylinder, four-
stroke, air-cooled direct injection diesel engine (Palmera, model PA-
HP170F). Table 1 presents the main specifications of the test engine.
The experimental system was produced by TecQuipment (model
TD114) and its photograph was given in Fig. 1. Exhaust gases compo-
sition was determined by DiGas 4000 gas analyzer produced by AVL.
The accuracies of CO2 and CO are within ± 0.1% vol. and ± 0.01%
vol., respectively. Also, the accuracies of HC and NOx measurement are
within 1 ppm.
2.2. Ammonia addition system
In the present experimental study, for adding appropriate amounts
of ammonia solution into inlet air an elementary carburettor, which gas
and air throttles were dismantled and the other auxiliary equipments
were left out of order, was mounted on the inlet manifold of the test
engine. The air inlet of this carburetor was connected to the air con-
sumption measuring box by using a flexible hose. Furthermore, am-
monia solution flow rate was controlled by a fine threaded adjustment
screw which can change the carburetor main fuel jet section. Fig. 2
presents photograph of the adapted carburettor.
2.3. Operating conditions
In the present study; a new engine, produced for agricultural ap-
plications, was mounted in the experimental system. It is well known
Table 1
Main technical specifications of the test engine.
Engine Palmera PA-HP170 F direct injection diesel engine
Displacement 211 cm3
Number of cylinder 1
Bore & stroke 70 & 55 mm
Compression ratio 20: 1
Maximum power 3.1 kW @ 3600 rpm
Z. Şahin et al. Fuel 214 (2018) 330–341

Fig. 1. The photograph of the test system. (1-Test engine,

2-Measuring panel, 3-Exhaust gas analyzer).

NDF and ammonia solution are summarized. The details of the calcu-
lation process can be found in various references such as [19]. In the
present study, fuel consumption of the engine was determined by using
a scaled glass bulb and consumption duration of 50 mL of diesel fuel
was measured. By this way, effective power output, total fuel con-
sumption, brake specific fuel consumption (bsfc) and effective effi-
ciency have been calculated by using the following relations.
Tb ω
Ne (kW) = 0,1013 T0/293Xhum
p0 (1)

Δmf ΔVρd 3600

B[kg/h]= = NDF
Δt Δt106 (2a)

Δmf (50ρd + VA ρA )3600

B[kg/h]= = AF
Δt Δt106 (2b)
Fig. 2. The photograph of adapted fumigation carburetor, which used to add ammonia in
B 3600
intake air. be [kg/kWh] = , ηe =
Ne QLHV be (3a,3b)

that there is a first running in process for a new engine. The producer
firm recommended that the test engine may not be operated under very
high loads during the running in process. The test engine was worked at
low engine speeds approximately 5 min during both before running in
process and actual experiments. Then, the engine was gradually loaded.
After the test engine was operated for 20 h of running in process, its oil
was changed. Then, the actual tests were stared.
In this study, the effects of AF on engine performance and exhaust
emissions were experimentally studied and compared for different
loads and speeds. Here, experiments were conducted at three different
engine speeds of 2200, 2600 and 3000 rpms for five different ammonia
solution ratios (2, 4, 6, 8 and 10%, by vol.). Here, engine loads of (2, 4,
6 and 8) Nm at 2200 and 3000 rpms and (2, 4, 6, 8 and 10) Nm at
2600 rpm were selected. Tests were firstly carried out for NDF to obtain
a database for comparison of the results of AF. After completed NDF
experiments, AF tests were performed. In the fumigation method, the
adapted carburetor was mounted on the inlet manifold of the test en-
gine and thus, ammonia solution was introduced into intake system.
Here; for fumigation tests, to obtain 5 different ammonia ratios of ∼(2,
4, 6, 8 and 10)%, by vol., carburetor main jet opening was adjusted at 5
different position.
2.4. Calculation of engine performance characteristics and determination
fumigated fuel properties
In this section, the principles of the calculation of engine perfor-
mance parameters and determination of fumigated fuel properties for
In Eq. (1); Tb (Nm) is brake torque, ω is angular velocity of the
crankshaft, p0 (MPa) and T0 (K) are pressure and temperatures of am-
bient air, respectively. Xhum is the humidity correction factor and it is
determined depending on dry and wet thermometer temperatures. In
Eqs. (2a); ΔV is the volume of consumed fuel, Δt (s) is the duration of
consumption of ΔV volume (50 mL) of fuel, ρd is the density of diesel
fuel, and LHV is the lower heating value of diesel fuel or diesel fuel-
ammonia solution mixture. Here, lower heating values of diesel fuel and
ammonia solution have been calculated by using well known Mende-
leyev formula [20].
QLHV [kJ/kg] = [33.91c′ + 125.6h′−10.89(oy ′−s′)−2.51(9h′−w)]10−3 (4)
In Eq. (3), c′, h′, oy′, s′andw′ represent elemental composition of fuel,
and their values for diesel fuel and ammonia solution have been given
in Table 2. For fumigation case, lower heating values have been de-
termined by using the following relation given by Durgun [19,21].
n
∑ (xiρi LHV)
i= 1 x dρd LHVd + xA ρA LHVA
LHVAF (kJ/kg) = n =
x dρd + xA ρA
∑ (xiρi )
i= 1 (5)
where ρd and ρa are the densities of diesel fuel and ammonia solution
respectively, and xd and xA are the volumetric percentages of diesel fuel
and ammonia solution in the mixture, respectively.
In the present study, to see clearly the effects of AF on engine per-
formance and exhaust emissions, variation ratios of engine character-
istics and exhaust emissions in respect of NDF were calculated. For

332
Z. Şahin et al. Fuel 214 (2018) 330–341

Table 2 Si = ρi / ρd S1 = ρd / ρd = 1, S2 = ρa / ρd = 0.890/0.817 = 1.089

The main properties of diesel fuel and ammonia [7,8,22].
C1 is cost of diesel fuel, C2 is cost of ammonia and Δbe/be is dif-
Diesel fuel Ammonia
ference ratio of bsfc, ρa and ρd are densities of ammonia and diesel fuel,
Chemical formula C12.3H22.2 NH4OH respectively. Here, units of (C1, C2), (ρa, ρd) and be are (TL/lt), (kg/m3)
Molecular mass [kg/kmol] 170.1079 35.046 and (kg/kWh) respectively. In Trabzon, Turkey for 2016 currency, 1$ is
Density [kg/m3] 0.8165* 0.8903* equal to 3.50 TL. The costs and other main characteristics of diesel fuel
Lower heating value [kJ/kg] 42994.28** 9839**
and ammonia are given in Table 2.
Cetane number 45 –
Latent heat of evaporation [kJ/kg] 270 –
Boiling point [°C] 180–360 –
Flash point [°C] ≥55 – 2.6. Error analysis and uncertainties
Kinematic viskosity, at 20 °C, [mm2/s] 3.4 –
Composition, mass [%] c′ = 0.8685 n′ = 0.3997
h′ = 0.1315 h′ = 0.1438
In the present study, error analysis of the measured and derived
o′y = 0.4565 values was done and uncertainties were also determined by using Kline
Cost [TL/lt], April 2016 3.64 2.83 and Mc. Clintock’s method [23]. Here, for the reason of being each
3.50 $ = 1TL*** value measured three times, Student’s t-distribution has been applied to
Some physical features
the experimental data. At the end of the statistical analysis of the ex-
* Measured in laboratory. perimental data, uncertainty intervals of bsfc and effective efficiency
** Calculated from Mendeleyev formula. values are determined at the levels of (0.297–1.565)%. Also, un-
*** TL: Turkish Lira. certainty intervals of bsfc and effective efficiency values have been
determined at the levels of (2.196–8.975)% for selecting the probability
example, variation ratio of bsfc was computed as follows: level of 90% by taking into account Student’s distribution. From the
evaluations of these results, it can be stated that the probably un-
Δbe
× 100[%] = ((be,AF −be,d)/be,d)100 certainties in the measuring of the principle values and in the derived
be (6)
values would not results in significant effect on the uncertainties of the
where be,AF and be,d are bsfc values for fumigated fuel and diesel fuel, numerical results.
respectively.

2.5. Cost analysis 3. Results and discussion

In the present study, cost analysis has also been done by using the In the present study, aqueous ammonia solution (% 25 am-
practical relationship, which was proposed originally by Durgun [20]. monia + % 75 water blends) has been investigated experimentally by
Here, by using variation ratio bsfc in respect to NDF and the actual applying fumigation method in a small diesel engine. In this method,
prices of ammonia and diesel fuel, combined fuel cost variation for ammonia solution is introduced into intake air by using an elementary
fumigated is calculated and compared. For this purpose, the following carburetor and ammonia solution-air mixture is compressed during the
formula has been used. compression process. Then, diesel fuel injected into these mixtures at
n
the end of the compression process. Here, any change has not been
⎡x + ∑ Xr ⎤ made in original injection system of the test engine and the engine
ΔC C2−C1 ⎢ 1 i i
Δb ⎥ mainly operates due to diesel principle. Experimental results related to
100[%] = 100 = ⎢ 2 ⎛1 + e ⎞−1⎥ 100
⎜ ⎟
n
C1 C1 ⎢ ⎝ be ⎠ ⎥ the influences of the ammonia solution on bsfc, effective efficiency,
⎢ x1 + ∑ Xis i ⎥ total fuel cost and exhaust emissions (CO2, CO, HC and NOx) have been
⎣ 2 ⎦ (7)
presented and compared to NDF in various figures. Here, ammonia
where solution fumigation has been abbreviated as AF and in figures and in
ri = Ci /C1, r1 = C1/C1 = 1, r2 = C2 /C1 = 2.83/3.64 = 0.777 the following paragraphs the abbreviation of AF has been used.

Fig. 3. (a and b) Variations and variations of the variation ratios of bsfc versus different ammonia ratios under four different loads at 2200, respectively.

333
Z. Şahin et al.
3.1. Brake specific fuel consumption (bsfc) and effective efficiency
Fig. 3a and b show the variations and variations of the variation
ratios of bsfc versus to the ammonia ratios for four different engine
loads at 2200 rpm, respectively. As can be seen in Fig. 3(a and b), bsfc
decreases for ∼2% and ∼4% AF ratios at only 6 Nm load. However,
bsfc starts to increase for increasing ammonia ratios at this load. At
other applied loads and ammonia ratios bsfc increases. As ammonia in
the fuel mixture is increased, bsfc increases significantly for selected
loads. It is well known that the flame temperature of ammonia is lower
than that of diesel fuel [7,8,13,14]. Thus, as expected, the combustion
temperature values may reduce with increasing of the amount of am-
monia in the combustion chamber. Low cylinder temperatures may
produce incomplete combustion and higher bsfc. Furthermore, aqueous
ammonia solution (25% ammonia + 75% water blends, NH4OH) was
used in present study. Thus, most of the mixture consists of water. It is
known from studies in the literature that the introduction water into
intake air reduces the cylinder temperatures in diesel engine. This may
increase bsfc. For example, Sahin et al. investigated water injection into
intake air at specified ratios for different engine speeds in a diesel en-
gine and it has been seen that the water injection has slightly increased
bsfc [24]. The obtained maximum increase ratio of bsfc is 10.79% for
10.31% AF ratio, under 2 Nm at 2200 rpm. Similar results have also
been reported by earlier researcher [7,13].
Fig. 4a and b show the variations and variations of the variation
ratios of effective efficiency versus to the ammonia ratio for various
engine loads at 2200 rpm, respectively. It can be clearly seen from Eq.
(3b) that effective efficiency is simply the inverse of the product of bsfc
and lower heating value of fuel, so there is bound to be of a close in-
terrelation between them. It can be observed from Fig. 4(a and b) ef-
fective efficiency increases for selected ammonia ratios. However, in-
crement ratios of effective efficiency are significant at low ammonia
ratios. Similarly, the increase ratios of bsfc at low ammonia ratios are
also lower than that of high ammonia ratios. It can be seen in Table 2
that the lower heating value of ammonia is nearly 4 times smaller than
of diesel fuel, which produces good effective efficiency. This can be
explained as follows. At 2200 rpm; bsfc has been obtained as 0.2957
and 0.3004 kg/kWh for NDF and 4% AF at 8 Nm load, respectively. In
this case, when effective efficiency was calculated by using Eq. (4)b, it
can be obtained as 0.2832 and 0.2922 for NDF and 4% AF at 8 Nm load,
respectively. Furthermore, the lower heating value of NDF and% 4 AF
(% 4 ammonia-% 96 diesel fuel mixture, by vol.) can be calculated as
42,994.28 kJ/kg and 41,012.378 kJ/kg, respectively. Therefore, effec-
tive efficiency increases for AF although bsfc increases for AF.
Fig. 4. (a and b) Variations and variations of the variation ratios of effective efficiency versus different ammonia ratios under four different loads at 2200, respectively.
334
Fuel 214 (2018) 330–341
It can be seen in Fig. 5(a and b) bsfc increases under 2 and 4 Nm
loads at 2600 rpm. It is evident that increment ratios of bsfc are
dominant at high ammonia ratios for these loads. At other selected high
loads (6, 8 and 10 Nm) bsfc decreases. On the other hand, it increases
slightly for 8 and 10% ammonia ratios at 6 Nm load. Particularly, an
improvement of bsfc has become very dominant at low ammonia ratios
as shown in Fig. 5b. As mentioned earlier, higher ammonia ratios re-
sults in incomplete combustion due to lower flame temperature of
ammonia.
Furthermore, AF at low loads may not be approaching its flamm-
ability limits and the premixed ammonia-air mixture in the cylinder can
be too lean to burn effectively. Therefore, decrement ratios of bsfc at
low loads are lower than that of at high loads [7,13]. The obtained
maximum decrease ratio of bsfc is 8.28% for 5.48% ammonia ratio
under 8 Nm load at 2600 rpm. The test engine is designed for agri-
cultural pumping applications and the most stable operation speed of
the engine is 2600 rpm. For this reason, the influences of ammonia on
engine performance and exhaust emissions at this engine speeds have
been more effective.
Fig. 6(a and b) show variations and variations of the variation ratios
of effective efficiency versus to ammonia ratios for five different engine
loads at 2600 rpm, respectively. Effective efficiency increases for se-
lected ammonia ratios and loads. It can be seen in Fig. 6b that increase
ratios at high loads are higher than that of low loads. Both the decrease
in bsfc and the reduction in the lower heating value of ammonia-diesel
fuel mixtures for AF increase the effective efficiency. The obtained
maximum increase ratio of effective efficiency becomes 15.02% for
8.50% ammonia ratio and 10 Nm at 2600 rpm.
Fig. 7(a and b) show variations and variations of the variation ratios
of bsfc versus to ammonia ratios for four different engine loads at
3000 rpm, respectively. bsfc increases for selected ammonia ratios and
loads. The obtained maximum increase ratio of bsfc is 8.49% for
10.07% ammonia ratio under 8 Nm at 3000 rpm. Variations and var-
iations of the variation ratios of effective efficiency versus to ammonia
ratios for four different engine loads at 3000 rpm are shown in Fig. 8(a
and b). Effective efficiency generally tends to increases for selected
ammonia ratios and loads. As can be seen from the inspection of
Fig. 8(a and b) the maximum increase ratio has been obtained as
11.54% for 9.67% ammonia ratio at 2 Nm.
3.2. Fuel economy
Variations of variation ratios of total fuel cost for the ammonium
fumigation compared to NDF for different loads at 2200, 2600 and
Z. Şahin et al. Fuel 214 (2018) 330–341

Fig. 5. (a and b) Variations and variations of the variation ratios of bsfc versus different ammonia ratios under five different loads at 2600, respectively.

Fig. 6. (a–b) Variations and variations of the variation ratios of effective efficiency versus to ammonia ratios under five different engine loads at 2600 rpm, respectively.

Fig. 7. (a and b) Variations and variations of the variation ratios of bsfc versus different ammonia ratios under four different loads at 3000, respectively.

335
Z. Şahin et al. Fuel 214 (2018) 330–341

Fig. 8. (a-b) Variations and variations of the variation ratios of effective efficiency versus to ammonia ratios under four different engine loads at 3000 rpm, respectively.

Fig. 9. (a–c) Variations of the variation ratios of cost versus to ammonia ratios under different engine loads at 2200, 2600 and 3000 rpms, respectively.

336
Z. Şahin et al.
Fig. 10. (a and b) Variations and variations of the variation ratios of CO2 versus different ammonia ratios under four different loads at 2200, respectively.
3000 rpms were presented in the Fig. 9(a–c), respectively. In the pre-
sent study the used ammonium solution and diesel fuel cost are
2.83 TL/L and 3.64 TL/L respectively. As can be seen from Fig. 9(b) that
the total cost of fuel for AF generally takes lower values than that of
NDF at 2600 rpm. However, the total cost of fuel generally increases at
2200 rpm and it is clear that as the ratios of ammonia are increased, the
total cost of the fuel increases at this engine speeds. Although the
changes are not very consistent, the total cost of fuel generally increases
at 3000 rpm. In short, the use of ammonia seems not to be economic at
2200 and 3000 rpms.
3.3. Exhaust emissions
In the present study, carbon dioxide (CO2), carbon monoxide (CO),
total hydrocarbon (HC), and nitrogen oxides (NOx) were measured for
various AF ratios, under four loads at 2200 rpm. However, these
emissions were measured for various AF ratios, under only 6 Nm load at
2600 rpm.
Fig. 10(a and b) present variations and variations of variation ratios
of CO2 for different AF ratios at 2200 rpm, respectively. It can be seen
these figures that CO2 percentage generally decreases for selected AF
ratios. It is known that ammonia is a carbon-free fuel [7,13], so it is
used as additional fuel, combustion produces lower CO2 than that of
Fig. 11. (a and b) Variations and variations of the variation ratios of CO versus different ammonia ratios under four different loads at 2200, respectively.
337
Fuel 214 (2018) 330–341
petroleum based fuels. This is a predictable result for using of ammonia
as an additive fuel.
Fig. 11(a and b) show variations and variations of variation ratios of
CO emissions for different AF ratios at 2200 rpm, respectively. CO
emissions generally increase for selected AF ratios and loads, expect for
8 Nm load. They decrease for all selected AF ratios under 8 Nm load.
Similarly, under 2 Nm load, CO emissions for AF were very similar to
that with NDF for (2 and 4)% AF ratios and they decrease for (6 and
8)% AF ratios. Similar results were found in literature [7,10,13]. A
possible explanation for the higher CO emissions is attributed to the
incomplete combustion due to low combustion temperature of am-
monia, especially for low loads [10].
As can be seen in Fig. 12b that variation ratios of HC emissions are
not consistent. Variations of HC for approximately up to 6% AF ratios at
selected loads are nearly similar to NDF. After this AF ratio, it generally
shows an increment tendency. As it is mentioned earlier, the flame
temperature of ammonia is lower than that of diesel fuel [7,13]. This
may result in lower cylinder temperature and pressure values. Thus, the
reduced temperature values cause a high level of HC emissions. Com-
bustion mechanisms which may occur in cylinder might be effective on
HC production. Flame quenching on the walls and the ammonia-air
mixture trapped in the piston-ring crevice may account for the main
source of unburned ammonia emissions [7,13]. The similar results have
Z. Şahin et al.
Fig. 12. (a and b) Variations and variations of the variation ratios of total hydrocarbon versus different ammonia ratios under four different loads at 2200, respectively.
also been reported by earlier researcher [7,8].
As can be seen in Fig. 13(a and b) that NOx emissions increase
significantly. It can be thought that the cylinder temperature values are
reduced for AF. However, NOx emissions increase. The possible reason
of this is the abundant fuel-bound nitrogen in ammonia. Furthermore,
excess air coefficient for AF ratios increases. This causes an increase in
NOx emissions [25]. Numerical values of excess air coefficients for NDF
and different AF ratios at 4 Nm load can be given as an example. The
excess air coefficient for NDF and (1.77, 4.40, 5.30, 7.41 and 10.71)%
AF ratios have been calculated as 4.236, 4.310, 4.296, 4.314, 4.317 and
4.280, respectively. The obtained maximum increase ratio of NOx
emission is 23% for 10.71% AF ratio and 4 Nm at 2200 rpm.
Variations and variations of the variation ratios of CO2, CO, HC, and
NOx for different ammonia ratios, under only 6 Nm load are shown in
Figs. 14–17 at 2600, respectively. Fig. 14(a and b) indicate variations
and variations of the variation ratios of CO2 for different AF ratios for
only 6 Nm load, respectively. CO2 increases for different AF ratios.
However, increment ratios of CO2 decreases with increasing AF ratios.
Actually, it is expected to decrease CO2 with the addition of ammonia.
However, it increases. The test engine operates very consistently at this
engine speed and it is considered that ammonia addition improves the
combustion process, which produces more complete combustion. Thus,
CO2 increases due to improvement of combustion. However, it starts to
Fig. 13. (a and b) Variations and variations of the variation ratios of NOx versus different AFratios under four different loads at 2200, respectively.
338
Fuel 214 (2018) 330–341
decrease with increasing AF ratios. Because ammonia is a carbon-free
fuel and this property becomes very dominant than improvement
combustion for higher AF ratios.
Figs. 15(a and b) and 16(a and b) clearly indicate that CO and HC
emissions increase significantly for all of AF ratios at this engine speed.
As explained earlier, this can also be explained by incomplete com-
bustion. Also, Subramanian showed that water injection into intake air
increases HC emissions and he explained possible increment reason as
follows: The water is closely in contact with the diesel particles and
quenches the combustion process. The used ammonia solution results
are similar to the results of water addition into intake air. Similar re-
sults for ammonia and water injection into the intake air have also been
reported by other researchers [7,13,26].
As can be seen in Fig. 17(a and b), NOx emissions increase con-
siderably with increasing AF ratios. It is supposed that the increase in
NOx emissions is primarily due to fuel NOx formation by used ammonia
solution.
3.4. The deteriaration effects of ammonia on carburetor and carburetor
elements
A simple carburetor has been used to introduce ammonia into intake
air as shown in Fig. 2. Ammonia has caused to damage some elements
Z. Şahin et al. Fuel 214 (2018) 330–341

Fig. 14. (a–b) Variations and variations of the variation ratios of CO2 versus to ammonia ratios under 6 Nm load at 2600 rpm, respectively.

Fig. 15. (a–b) Variations and variations of the variation ratios of CO versus to ammonia ratios under 6 Nm load at 2600 rpm, respectively.

Fig. 16. (a and b) Variations and variations of the variation ratios of HC versus different ammonia ratios under 6 Nm load at 2600, respectively.

339
Z. Şahin et al.
Fig. 17. (a and b) Variations and variations of the variation ratios of NOx versus different ammonia ratios under 6 Nm load at 2600, respectively.
Fig. 18. (a) Deterioration of the carburetor elements (b) Residue and deteriaration on the
inner surface of carburetor.
of carburetor. In fact, the adjustment screw which can change the
carburetor main fuel jet section has been deformed. The deteriaration
effects of ammonia on carburetor and carburetor elements present in
Fig. 18(a and b). Also, as can be seen in Fig. 18b that ammonia has
caused some residues on the carburetor inner surfaces.
4. Conclusions
In the present study the effects of ammonia solution addition into
intake air on performance, fuel cost and emission characteristics were
investigated experimentally and compared with that of NDF in a small
DI diesel engine installed at the authors’ laboratory. Based on the ex-
perimental results the main effects of the ammonia solution addition
can be summarized as follows:
1. The ammonia addition into the intake manifold increases the bsfc
340
Fuel 214 (2018) 330–341
under selected four different loads at 2200 and 3000 rpms. The
obtained maximum increase ratio of bsfc is 10.79% for 10.31% AF
ratio, under 2 Nm at 2200 rpm and 8.49% for 10.07% AF ratio,
under 8 Nm at 3000 rpm, respectively. bsfc generally decreases for
selected ammonia ratios and loads at 2600 rpm. At this engine speed
the maximum decrement ratio of bsfc is determined as 7.28% for
5.48% AF ratios, under 8 Nm load. The best results in terms of en-
gine performance and exhaust emissions for ammonia solution fu-
migation have been obtained at 2600 rpm because of the test engine
is designed to operate consistently at this engine speed.
2. Ammonia solution addition into the intake manifold generally in-
creases the effective efficiency for selected loads at 2200, 2600 and
3000 rpms. The obtained maximum increase ratio of effective effi-
ciency is 15.05% for 8.5% AF ratio, under 8 Nm load at 2600 rpm.
3. The effects of different ammonia solution fumigation ratios on ex-
haust emissions have experimentally been investigated for (2, 4, 6
and 8) Nm loads at 2200 rpm, under only 6 Nm load at 2600 rpm.
CO and HC emissions generally tend to increase for ammonia so-
lution fumigation. Also, ammonia solution fumigation significantly
increases NOx emissions. For 11% ammonia solution fumigation
ratio on average 23% increment in NOx emission have been de-
termined for 4 Nm. At this engine speed, CO2 decreases for ammonia
solution fumigation ratios.
At 2600 rpm, the ammonia solution fumigation significantly in-
creases CO, HC and NOx emissions and it increases CO2. However,
increment ratios of CO2 decrease with increasing ammonia ratios.
4. The total cost of fuel for AF generally takes lower values than that of
NDF at 2600 rpm. However, it generally increases at 2200 and
3000 rpms.
5. Ammonia solution has caused to damage some elements of car-
buretor and form some residues on the carburetor inner surfaces.
6. Overall, this experimental study shows that ammonia can be used as
an alternative additive fuel for diesel engines. However, more
comprehensive theoretical and experimental studies related to using
of ammonia solution and ammonia in diesel engines need to be
performed.
Acknowledgement
We would like to present our thanks to Karadeniz Technical
University (KTU) Engineering Faculty which support our study finan-
cially and Mechanical Engineering Dep. Internal Comb. Engine Lab.
Technician Şenol DÜBÜŞ who has assisted to the experiments.
Z. Şahin et al.
References
[1] Yao M, Wang H, Zheng Z, Yue Y. Experimental study of n-butanol additive and
multi-injection on HD diesel engine performance and emissions. Fuel
2010;89:2191–201.
[2] Giakoumis EG, Rakopoulos CD, Dimaratos AM, Rakopoulos DC. Exhaust emissions
with ethanol or n-butanol diesel fuel blends during transient operation: a review.
Renewable Sustainable Energy Rev 2013;17:170–90.
[3] Rakopoulos DC, Rakopoulos CD, Giakoumis EG, Dimaratos AM, Kyritsis DC. Effects
of butanol-diesel fuel blends on the performance and emissions of a high-speed DI
diesel engine. Energy Convers Manage 2010;5:1989–97.
[4] Bae C, Kim J. Alternative fuels for internal combustion engines. Proc Combust Inst
2017;36:3389–413.
[5] Goldsworthy L. Fumigation of a heavy duty common rail marine diesel engine with
ethanol–water mixtures. Exp Thermal Fluid Sci 2013;47:48–59.
[6] Brohi EA, Ammonia as fuel for internal combustion engines? An evaluation of the
feasibility of using nitrogen-based fuels in ICE, master’s thesis in sustainable energy
systems, Department of Applied Mechanics, Chalmers University of Technology,
2014.
[7] Reiter AJ, Kong SC. Combustion and emissions characteristics of compression-ig-
nition engine using dual ammonia-diesel fuel. Fuel 2011;90:87–97.
[8] Ryu K, Zacharakis-Jutz GE, Kong SC. Effects of gaseous ammonia direct injection on
performance characteristics of a spark ignition engine. Appl Energy
2014;116:206–15.
[9] Ryu K, Zacharakis-Jutz GE, Kong SC. Performance enhancement of ammonia-fueled
engine by using dissociation catalyst for hydrogen generation. Int J Hydrogen
Energy 2014;5:2390–8.
[10] Ryu K, Zacharakis-Jutz GE, Kong SC. Effects of fuel compositions on diesel engine
performance using ammonia-DME mixtures, SAE Technical Paper No: 2013–01-
1133, 2013.
[11] Saravanan N, Nagarajan G. Experimental investigation of hydrogen port fuel in-
jection with SCR in a DI diesel engine, SAE Technical Paper No: 2009–26-033,
2009.
[12] Gill SS, Chatha GS, Tsolakis A, Golunski SE, York APE. Assessing the effects of
partially decarbonising a diesel engine by co-fuelling with dissociated ammonia. Int
341
Fuel 214 (2018) 330–341
J Hydrogen Energy 2012;37:6074–83.
[13] Reiter AJ. Combustion and emissions characteristics of compression-ignition engine
using dual ammonia-diesel fuel. Iowa: Master of Science, Iowa State University;
2009.
[14] Starkman ES, James GE, Newhall HK. Ammonia as a diesel engine fuel: theory and
application. Soc Automot Eng 1968;1968(76):3193–212.
[15] Feng T, Lu L. The characteristics of ammonia storage and the development of
model-based control for diesel engine urea-SCR system. J Ind Eng Chem
2015;28:97–109.
[16] Zhao J, Che, Z, Hu Y, Chen H. Urea-SCR process control for diesel engine using.
[17] Abu-Qudais M, Haddad O, Qudaisat M. The effect of alcohol fumigation on diesel
engine performance and emissions. Energy Convers Manage 2000;41:389–99.
[18] Sahin Z, Durgun O, Kurt M. Experimental investigation of improving diesel com-
bustion and engine performance by ethanol fumigation-heat release and flamm-
ability analysis. Energy Convers Manage 2015;89:175–87.
[19] Durgun O. Experimental Methods in Engines. Lecturer notes for laboratory.
Karadeniz Technical University, Engineering Faculty, Mechanical Eng. Dep.; 1990.
[20] Kolchin A, Demidov V. Design of automotive engines. Moscow: Mir Publishers;
1984.
[21] Durgun O, Ayvaz Y. The use of diesel fuel–gasoline blends in diesel engines. In: Proc
of the First Trabzon International Energy and Environmental Sym.; 1996. p.
905–912.
[22] Zamfirescu C, Dincer I. Using ammonia as a sustainable fuel. J Power Sources
2008;185:459–65.
[23] Holman JP. Experimental methods for engineers. 7th ed. New York: McGraw-Hill;
2001.
[24] Sahin Z, Tuti M, Durgun O. Experimental investigation of the effects of water
adding to the intake air on the engine performance and exhaust emissions in a DI
automotive diesel engine. Fuel 2014;115:884–95.
[25] Heywood JB. Internal combustion engine fundamentals. New York: McGraw-Hill;
1998.
[26] Subramanian KA. A Comparison of water-diesel emulsion and timed injection of
water into the intake manifold of a diesel engine for simultaneous control of NO and
smoke emissions. Energy Convers Manage 2011;52:849–57.